Mentavo
"Why did you take that flat?"
The answer comes immediately: the layout, the location, the light. It sounds considered, and it probably is.
Whether it describes what actually tipped the decision is another question – and harder to answer than the confidence of the reply suggests.
What introspection actually reaches
The distinction at issue is narrow and consequential.
Looking inward does turn something up: a thought, a sensation, a leaning towards one option. That access is real.
What can't be observed is the process beforehand – what led to this particular thought appearing rather than another. That part runs without an audience.
So when asked for the reason, there's no transcript available. What happens instead is reconstruction: a plausible account assembled from what is known.
And because it's built from familiar material, it doesn't feel like a construction. It feels like remembering.
What the research shows
This field has been studied for decades, usually with a similar setup. People's behaviour is influenced by a factor they don't notice – the order in which things are presented, for instance. Then they're asked why they chose what they chose.
The answers generally don't name the actual influence. They name other, reasonable-sounding factors – and they're delivered with conviction.
Related to this is work on what's called choice blindness. Under certain experimental conditions, when people are shown an option other than the one they picked and told it was their choice, not everyone notices – and some then go on to justify a choice they never made.
That sounds more dramatic than it is. Effects like these appear under particular conditions, not universally. But they show how readily the explaining starts.
Where self-report works well – and where it doesn't
The obvious conclusion would be to distrust your own account in general. That would be wrong, and it isn't what the findings suggest.
People are right in plenty of cases, particularly when the influence was clear and present. Someone who says they cancelled because they were ill is describing it accurately.
It gets uncertain when the influences are quiet.
| Question | Self-report fairly reliable | Self-report less reliable |
|---|---|---|
| Why cancel today? | a clear, nameable reason | vague reluctance with no obvious cause |
| Why this option? | few, clearly different options | many similar options in a sequence |
| Why so irritable? | a conflict immediately beforehand | sleep, hunger, noise, time of day |
The right-hand column describes exactly the cases where people explain most confidently – because the absence of an obvious reason intensifies the search for one.
Why this matters for self-reflection
Here's the real consequence, and it isn't discouraging.
Thinking in order to find a reason reliably produces one. The question "why do I react like this?" almost always yields an answer – regardless of whether it's correct.
That's why pure reflection often stalls on recurring topics: it generates a fitting explanation every time, and each one seems convincing in the moment. Why that still doesn't produce change is covered in the piece on self-awareness without change.
One way out lies in the order of operations: collect first, interpret second.
Separating observation from explanation
An observation is something you can note down without interpreting: when it was, what preceded it, how long it lasted.
An explanation is the sentence that makes sense of it.
Keeping the two apart sounds pedantic and changes a great deal in practice, because observations stay comparable over time and explanations don't. The same distinction is applied to everyday thinking in thought or fact.
In practice this means the explanation may stand alongside – as one of several, marked as a hypothesis rather than a finding.
What several observations can do
A single case can always be explained. Twenty cases contradict an explanation that doesn't hold.
That's exactly the value of recorded observations: they existed before the interpretation. Writing down over weeks how a day went leaves you with a basis that doesn't bend to today's mood – and that occasionally contradicts a favourite explanation.
In Mentavo, daily check-ins build up that kind of run, and the Mentavo Map shows which areas frequently occur together. It explicitly doesn't show why something connects – it supplies the observations a hypothesis can be built from in the first place.
At what point a cluster may count as repetition at all is covered in pattern or coincidence.
The question that gets you further
If "why?" reliably produces an answer that can't be checked, the improvement lies in asking something else.
Useful questions aim at circumstances rather than motives: when does it happen? What came immediately before? In which situations does it not happen?
The difference isn't a matter of phrasing. "Why am I so short with my sister?" invites an explanation about the relationship. "On which occasions have I been short-tempered in recent weeks?" invites a list – and the list can surprise you, for instance if three of the four fell after short nights.
Circumstances can be checked; motives can't. That doesn't make the second question deeper, but it makes it more load-bearing.
It also has a practical side effect: looking for circumstances turns up something changeable more often than looking for character traits does.
What this article doesn't claim
It doesn't claim that people have no access to their motives. Many explanations are accurate, and the capacity to understand your own behaviour is real.
Nor does it claim that after-the-fact explanations are generally wrong. A reconstruction can be correct – it just isn't evidence for itself.
How far these findings extend remains a live question in the research, particularly how well laboratory conditions carry over to ordinary decisions. The core observation – that explaining starts easily and feels certain – is much less contested.
And none of this is an argument for self-distrust. Doubting every statement you make about yourself only relocates the uncertainty.
In short: the explanation is a hypothesis
That a reason arrives quickly and sounds certain tells you very little about whether it's right.
Treating it as a hypothesis, and recording alongside it what was actually observable, costs nothing – and after a few weeks leaves you with something that can confirm the hypothesis or undo it.
